250g-500g 1/0.5mm Lötzinn Lötdraht Löt mit Flussmittel Lötkolben Lötstation DHL (WHL #68F1)
(Follow-up post to the solder test of WHL #68)
While we’re at it – I couldn’t resist testing the flux percentage. Which also resulted in having large samples for a visual comparison. Contrary to my expectations, the lead-free solder wasn’t the one that stood out…
First things first. New batch, same naming scheme as before. Since I already measured weight per meter, I just picked one sample length from a (new) lead-free U1.0 wire and cut the others to expected length. 100cm of that stuff is 5.42(196)g, for comparison.
Yes, one could correct the previous data with the new measurements (more wire – more precise data), but I didn’t really care all that much about precision, it just needed it to roughly match up. I think it is advantageous to have sample pills of the same weight so that one can already judge from the size of the blob. Which was, by the way, created on a stainless steel foil using a hot air gun, since that minimizes cross-contamination via the soldering iron tip. Except for the China solder, all others behaved really well and did not create any tiny specks that wouldn’t easily unite with the main blob and skew measurements this way. And that was not caused by some chemical reaction or the mixture of all the fluxes and their residues, the China solder was No. 3 and the F0.7 was produced last. Worked nicely on the really burnt/coated surface, so there’s that.
All solder was fully cooled down and then cleaned multiple times with isopropyl alcohol, so that there should be no flux present. Oxides however is a different story, but they should only cover the surface and not play a major role in such large samples – more oxide coating would reduce apparent flux content as they add weight from captured oxygen that wasn’t present in the beginning. Again, except for the China crap all solder wires behaved reasonably well and had a highly rounded surface anyway, so there are no large reservoirs to begin with. An unfortunate crease on the bottom side of the lead-free sample being a small exception to this. The wrinkles on top are an indicator for the coefficient of thermal expansion of each material mix, since those only appeared when the outer shells were already solid (partially crystallized!), with the center barely above solidus temperature filling the gaps. Pure lead has an expansion coefficient of 28.9µ/K (at 20°C) while tin has around 3/4 of that, 22.0µ/K to be precise. So for a first rough guess, large amounts of leaded solder will show deeper recesses on uneven cooling than lead-free solder which is usually 95%+ tin, so there’s much less material that can have a different coefficient in the first place. Always keep in mind that higher (or lower) values aren’t better (worse) per se, it always needs to be compared to the material system it is bonding with. Solder compounds for ceramics and glasses are tailored to match these materials over a wide range of temperatures, while regular electronics solder will just do the job and the magic smoke will escape way before thermal expansion of solder will have any effect. (Repeated thermal cycling is a different story)
Back to the four contestants. From left to right:
F0.5: 8669.15mg sample (raw wire)
F0.7: 8657.32mg sample
U1.0: 8666.22mg sample
C0.5: 8648.55mg sample
F0.5: “2.5%” flux, 8466.70mg final value. That’s 97.66% residual weight, indicating 2.34% flux.
F0.7: “3.5%” flux, 8379.37mg final value. 3.21% flux.
U1.0: (unknown brand, type, flux amount) 8438.63mg final value. 2.63% flux. Could be 2.5%, could be 3%. Reasonable amount for lead-free stuff since that usually needs more (and more aggressive) flux to work properly.
China C0.5… “2.0%” flux: 8517.58mg final value. 1.51% real flux amount.
Can’t say that’s a surprise, really…
Note also the different volumes – one can easily eyeball the lower density = higher volume of the lead-free sample, as well as the higher density = smaller volume of the China solder. That one also has a horrific surface tension as discussed previously, leading to a flattened appearance and, most notably, no real droplet shape. This is extremely undesirable for a solder compound both for mechanical stability as well as ease of inspection. This crap basically has cold solder joints by default.
The flux values could now be used to recalculate density and to make a better guess of compound composition – that’s up to the reader for exercise. It should downgrade overall density without flux to around 9.6 to 9.7g/cm³, making it more of a 30-70 type solder than the previously estimated 25-75. Both ways, it’s way off the 60-40 it should be, making this, once again, a very poor choice for electronics solder.